Skip to main content

Table 2 Evaluation of tandem repeat analysis methods on the ABCA7 VNTR

From: NanoSatellite: accurate characterization of expanded tandem repeat length and sequence through whole genome long-read sequencing on PromethION

Method

Accuracy (%)

Relative standard deviation (%)

Number of spanning reads

Expanded read detection (%)

Sequence composition

Albacore + tandem-genotypes

67.3

36.0

154

63

Low consistency

Scrappie events + tandem-genotypes

83.3

14.8

177

88

Low consistency

Scrappie raw + tandem-genotypes

87.7

5.5

181

25

Low consistency

Guppy “flip-flop” + tandem-genotypes

91.2

3.1

176

75

Low consistency

NanoSatellite

90.5

5.6

194

100

High consistency

  1. Accuracy corresponds to the degree of resemblance of the average length estimation and Southern blotting length. The relative standard deviation depicts the spread of length estimates to the mean. The total number of ABCA7 VNTR spanning reads detected per method is shown under “Number of spanning reads”; shown in more detail in Additional file 1: Table S2. “Expanded read detection” corresponds to the proportion of expanded reads that were detected using the accompanying method compared to the method that detected most expanded reads, with 100% corresponding to the detection of all expanded reads. Sequence composition is based on the resemblance between base called sequences and the ABCA7 VNTR reference sequence for conventional tools (Additional file 1: Table S1), and the reoccurrence of alternative TR unit patterns in independent sequencing reads for NanoSatellite (Fig. 4)